Research map / 3D vision / Face and shape
Relightable Holoported Characters: Capturing and Relighting Dynamic Human Performance from Sparse Views
Kunwar Maheep Singh, Jianchun Chen, Vladislav Golyanik, Stephan J. Garbin, Thabo Beeler, Rishabh Dabral, Marc Habermann, Christian Theobalt
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| use RelightNet with physics-informed features instead of baselines | HPC + NG · Subjects 1-5 relighting | PSNR: S1 30.52 -> 31.38; S3 29.80 -> 29.09 | Mixed |
| train on relit frames with ground-truth environment maps | R4D · Subject 1 relighting | PSNR: 26.09 -> 29.89 | Improved |
| add geometry features (ablation) | RHC · Ablation (one subject) | PSNR: 31.73 -> 32.07 | Improved |
| add albedo feature (ablation) | RHC · Ablation (one subject) | LPIPS (x100): 5.78 -> 5.55 | Improved |
| add diffuse shading feature (ablation) | RHC · Ablation (one subject) | PSNR: 31.59 -> 32.07 | Improved |
| add camera encoding (ablation) | RHC · Ablation (one subject) | PSNR: 31.52 -> 32.07 | Improved |
| use cross-attention for environment map instead of concatenation (ablation) | RHC · Ablation (one subject) | PSNR: 31.88 -> 32.07 | Improved |
| add coarse normal stack (ablation) | RHC · Ablation (one subject) | PSNR: 31.96 -> 32.07 | Improved |
| use dense-view skeleton tracking instead of sparse views (ablation) | RHC · Ablation (one subject) | PSNR: 30.72 -> 32.07 | Improved |
| use 4 input views instead of 0 (pose only) (ablation) | RHC · Ablation (one subject) | PSNR: 31.39 -> 32.07 | Improved |
| use 4 input views instead of 2 (ablation) | RHC · Ablation (one subject) | PSNR: 32.00 -> 32.07 | Improved |
| interleaved env-map capture instead of OLAT capture (ablation) | RHC · Ablation (one subject) | PSNR: 26.42 -> 32.07 | Improved |
Closest papers by meaning
- Comprehensive Relighting: Generalizable and Consistent Monocular Human Relighting and Harmonization (CVPR 2025 · similarity 0.89)
- ROGR: Relightable 3D Objects using Generative Relighting (NeurIPS 2025 · similarity 0.88)
- Relightable and Animatable Neural Avatar from Sparse-View Video (CVPR 2024 · similarity 0.87)
- UniLumos: Fast and Unified Image and Video Relighting with Physics-Plausible Feedback (NeurIPS 2025 · similarity 0.87)
- Holoported Characters: Real-time Free-viewpoint Rendering of Humans from Sparse RGB Cameras (CVPR 2024 · similarity 0.87)
- UniRelight: Learning Joint Decomposition and Synthesis for Video Relighting (NeurIPS 2025 · similarity 0.87)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.
